Figure 2 - uploaded by Jihad Anwar
Content may be subject to copyright.
Download
View publication
16: The scale-space computation by SIFT and SURF. The left figure gives the computation as done by SIFT, the right figure as done by SURF. As can be seen in the image, SIFT computes the scale-space by varying the image size, while SURF computed the scale-space by varying the filter size.

16: The scale-space computation by SIFT and SURF. The left figure gives the computation as done by SIFT, the right figure as done by SURF. As can be seen in the image, SIFT computes the scale-space by varying the image size, while SURF computed the scale-space by varying the filter size.

Source publication
Figure 1.1: Some examples illustrating the difficulties in person...
Figure 2.1: Simple program for feature detection.
Figure 2.2: Simple program for matching feature.
Figure 2.3: Different region
+50 Figure 2.4: Original image for test.
Human detection and tracking in video Sequence
Thesis
Full-text available
  • Jul 2016
Humans detection and tracking in videos are difficult and challenging, owing to the motion of the subjects. It is a general task to many applications, such as video surveil-lance and security, intelligent vehicles, etc. In this regard, our thesis introduces several visual methods for detecting and tracking humans’ body and their scope. In this thesi...
Cite
Download full-text

Similar publications

The scatter of TN rate while learning of Haar features for different...
Shapes of Haar descriptors used for training. The number of internal...
Examples of pedestrian and faces samples
Distributions of KL distances between positive and negative samples for...
+7 Distributions of KL distances between positive and negative samples,...
Haar descriptors preliminary sampling
Article
Full-text available
  • Jun 2022
The selection of meaningful features for training computer vision solutions is the most important part of a robust framework construction. A huge amount of possible features with different parameters makes this task chalenging. We have developed a preliminary procedure of sampling of Haar descriptors for more effective learning of the binary classi...
View